High quality non-blind image deconvolution using the Fields of Experts prior

Abstract

In this paper, we propose a non-blind image deconvolution method under Bayesian probabilistic framework. A robust prior, i.e., the Fields of Experts (FoE) prior which is learned from natural images is adopted to regularize the latent image. To solve the resulted optimization problem, the Split Bregman algorithm is utilized. We use a data set to verify the effectiveness of the proposed approach. Experimental results show that the restored images are of higher quality than that of some state of the art algorithms.